[DSP-10] Voltage readings and more

[Don AE5K]

I am beginning to feel like a dentist -- probing around amongst the 
surface mount devices looking for when it hurts.  The suggestion to 
probe with an insulated tool did show a sensitivity in the general area 
of U5.  The DSP-10 has been on for 36 hours now and power meter measured 
output has been +3.4 dBm this afternoon -- that is until I apparently 
disturbed something and it went down to +1.8 and a couple other low 
readings.  It's back to +3.4 now.

Further probing failed to make it jump again.  This gives me some hope 
since I've been seeing it in past days vary from -0.6 to +3.6 in sudden 
jumps. Visual inspection with a bright light and large magnifying glass 
did not show up any problems.

I also went through the various printed/web notes plus each and every 
email received yesterday, making every check suggested.

Before this dental probing with insulated tool (an old coil tuning 
tool), I did the following DC voltage measurements:

Unless otherwise noted:

Using QST 1999 schematic designations
CW mode in transmit but Key UP (no r.f. output)
Output pwr setting = 100
Input voltage = 13.6 vdc
+10T when in transmit mode = 10.5 vdc
+10T when in receive mode = 0.0 vdc

L6-R4-C13 junction = 0.91 vdc
R4 (+10T side) = 10.5 vdc
U4 pin 1 = 1.86 vdc
U4 pin 3 = 3.84 vdc
U5 pin 1 = 1.72 vdc
U5 pin 3 = 5.00 vdc
Q2 base (probed at C45) = 2.04 vdc
Q2 coll (probed at jct L18 & C50 = 10.5 vdc
R10/R11/C49 jct = 1.11 vdc

 From what I can read, figure out for myself, and have been told, all 
these voltages are in the ballpark.  It figures that the 2N5109 (Q2) has 
about 50 ma. emitter current if my Ohms law was correct.

I then went on to try with an r.f. probe, but at C41 it was not 
sensitive enough to move the meter when in tx and key down.  I would 
guess the level is down in the pre-conductive region of the diode.

I have never done this inter-stage checking before so correct me if I am 
wrong in my thinking.  I will take the KA7EXM meter without the 40 db 
tap, and since it has built in 50 ohm load, I should be able to use a 
short piece of coax with DC isolating capacitor to probe the inputs and 
outputs of each stage - U4 U5 and Q2.

And since each stage is supposedly in the neighborhood of 50 ohm 
impedance, and since I'm adding an additional 50 ohm load at the meter, 
I would expect to see approximately a reading of half the actual power. 
Absolute power is not necessary to detect an (almost) dead amplifier 
stage I don't believe.

Am I wrong in my assumptions?  If am wrong, suggest how I should go 
about checking each stage. I want to try to isolate the stage giving me 
the problem(s).  My plan then is the touch up the soldering each of the 
components used by that stage. (If heat sensitive R or C, then maybe it 
will make it fail altogether.)

As always, help is appreciated.  And with Bob coming out with new 
software and some of the old kits coming out of the closet, the DSP-10 
lives on!

Don AE5K